knowledge even of the course and extent of the current in this ocean. A thorough investigation of it with the improved methods of our time is therefore an inevitable necessity.

As the Gulf Stream is of so great importance to Northern Europe in general, but especially to us Norwegians, it was not a mere accident that three separate expeditions left Norway in the same year, 1910 -- Murray and Hjort's expedition in the Michael Sars, Amundsen's trial trip in the Fram, and Nansen's voyage in the gunboat Frithjof -- all with the object of investigating the conditions in the North Atlantic. The fact that on these three voyages observations were made approximately at the same time in different parts of the ocean increases their value in a great degree, since they can thus be directly compared; we are thus able to obtain, for instance, a reliable survey of the distribution of temperature and salinity, and to draw important conclusions as to the extent of the currents and the motion of the ma.s.ses of water.

Amundsen's trial trip in the Fram and Nansen's voyage in the Frithjof were made with the special object of studying the Gulf Stream in the ocean to the west of the British Isles, and by the help of these investigations it is now possible to chart the current and the extent of the various volumes of water at different depths in this region at that time.

A series of stations taken within the same region during Murray and Hjort's expedition completes the survey, and provides valuable material for comparison.

After sailing from Norway over the North Sea, the Fram pa.s.sed through the English Channel in June, 1910, and the first station was taken on June 20, to the south of Ireland, in lat. 50 50' N. and long. 10 15' W., after which thirteen stations were taken to the westward, to lat. 58 16' N. and long. 17 50' W., where the ship was on June 27. Her course then went in a northerly direction to lat. 57 59'

N. and long. 15 8' W., from which point a section of eleven stations (Nos. 15 -- 25) was made straight across the Gulf Stream to the bank on the north of Scotland, in lat. 59 88' N. and long. 4 44' W. The voyage and the stations are represented in Fig. 2. Temperatures and samples of water were taken at all the twenty-four stations at the following depths: surface, 5, 10, 20, 30, 40, 50, 75, 100, 150, 200, 300, 400, and 500 metres (2.7, 5.4, 10.9, 16.3, 21.8, 27.2, 40.8, 54.5, 81.7, 109, 163.5, 218, and 272.5 fathoms) -- or less, where the depth was not so great.

The Fram's southerly section, from Station 1 to 13 (see Fig. 3) is divided into two parts at Station 10, on the Porcupine Bank, south-west of Ireland. The eastern part, between Stations 1 and 10, extends over to the bank south of Ireland, while the three stations of the western part lie in the deep sea west of the Porcupine Bank.

[Fig. 2 and caption: Fig. 2. -- The "Fram's" Route from June 20 to July 7, 1910 (given in an unbroken line -- the figures denote the stations).

The dotted line gives the Frithjof's route, and the squares give five of the Michael Sars's stations.]

In both parts of this section there are, as shown in Fig. 3, two great volumes of water, from the surface down to depths greater than 500 metres, which have salinities between 35.4 and 35.5 per mille. They have also comparatively high temperatures; the isotherm for 10 C. goes down to a depth of about 500 metres in both these parts.

It is obvious that both these comparatively salt and warm volumes of water belong to the Gulf Stream. The more westerly of them, at Stations 11 and 12, and in part 13, in the deep sea to the west of the Porcupine Bank, is probably in motion towards the north-east along the outside of this bank and then into Rockall Channel -- between Rockall Bank and the bank to the west of the

[Fig. 3 and caption: Fig. 3. -- Temperature and Salinity in the "Fram's" Southern Section, June, 1910.]

British Isles -- where a corresponding volume of water, with a somewhat lower salinity, is found again in the section which was taken a few weeks later by the Frithjof from Ireland to the west-north-west across the Rockall Bank. This volume of water has a special interest for us, since, as will be mentioned later, it forms the main part of that arm of the Gulf Stream which enters the Norwegian Sea, but which is gradually cooled on its way and mixed with fresher water, so that its salinity is constantly decreasing. This fresher water is evidently derived in great measure directly from precipitation, which is here in excess of the evaporation from the surface of the sea.

The volume of Gulf Stream water that is seen in the eastern part (east of Station 10) of the southern Fram section, can only flow north-eastward to a much less extent, as the Porcupine Bank is connected with the bank to the west of Ireland by a submarine ridge (with depths up to about 300 metres), which forms a great obstacle to such a movement.

The two volumes of Gulf Stream water in the Fram's southern section of 1910 are divided by a volume of water, which lies over the Porcupine Bank, and has a lower salinity and also a somewhat lower average temperature. On the bank to the south of Ireland (Stations 1 and 2) the salinity and average temperature are also comparatively low. The fact that the water on the banks off the coast has lower salinities, and in part lower temperatures, than the water outside in the deep sea, has usually been explained by its being mixed with the coast water, which is diluted with river water from the land. This explanation may be correct in a great measure; but, of course, it will not apply to the water over banks that lie out in the sea, far from any land. It appears, nevertheless, on the Porcupine Bank, for instance, and, as we shall see later, on the Rockall Bank, that the water on these ocean banks is -- in any case in early summer -- colder and less salt than the surrounding water of the sea. It appears from the Frithjof section across the Rockall Bank, as well as from the two Fram sections, that this must be due to precipitation combined with the vertical currents near the surface, which are produced by the cooling of the surface of the sea in the course of the winter. For, as the surface water cools, it becomes heavier than the water immediately below, and must then sink, while it is replaced by water from below. These vertical currents extend deeper and deeper as the cooling proceeds in the course of the winter, and bring about an almost equal temperature and salinity in the upper waters of the sea during the winter, as far down as this vertical circulation reaches. But as the precipitation in these regions is constantly decreasing the salinity of the surface water, this vertical circulation must bring about a diminution of salinity in the underlying waters, with which the sinking surface water is mixed into a h.o.m.ogeneous volume of water. The Frithjof section in particular seems to show that the vertical circulation in these regions reaches to a depth of 500 or 600 metres at the close of the winter. If we consider, then, what must happen over a bank in the ocean, where the depth is less than this, it is obvious that the vertical circulation will here be prevented by the bottom from reaching the depth it otherwise would, and there will be a smaller volume of water to take part in this circulation and to be mixed with the cooled and diluted surface water. But as the cooling of the surface and the precipitation are the same there as in the surrounding regions, the consequence must be that the whole of this volume of water over the bank will be colder and less salt than the surrounding waters. And as this bank water, on account of its lower temperature, is heavier than the water of the surrounding sea, it will have a tendency to spread itself outwards along the bottom, and to sink down along the slopes from the sides of the bank. This obviously contributes to increase the opposition that such banks offer to the advance of ocean currents, even when they lie fairly deep.

These conditions, which in many respects are of great importance, are clearly shown in the two Fram sections and the Frithjof section.

The Northern Fram section went from a point to the north-west of the Rockall Bank (Station 15), across the northern end of this bank (Station 16), and across the northern part of the wide channel (Rockall Channel) between it and Scotland. As might be expected, both temperature and salinity are lower in this section than in the southern one, since in the course of their slow northward movement the waters are cooled, especially by the vertical circulation in winter already mentioned, and are mixed with water containing less salt, especially precipitated water. While in the southern section the isotherm for 10 C. went down to 500 metres, it here lies at a depth of between 50 and 25 metres. In the comparatively short distance between the two sections, the whole volume of water has been cooled between 1 and 2 C. This represents a great quant.i.ty of warmth, and it is chiefly given off to the air, which is thus warmed over a great area. Water contains more than 3,000 times as much warmth as the same volume of air at the same temperature. For example, if 1 cubic metre of water is cooled 1, and the whole quant.i.ty of warmth thus taken from the water is given

[Fig. 4. -- Temperature and Salinity in the "Fram's" Northern Section, July 1910]

to the air, it is sufficient to warm more than 3,000 cubic metres of air 1, when subjected to the pressure of one atmosphere. In other words, if the surface water of a region of the sea is cooled 1 to a depth of 1 metre, the quant.i.ty of warmth thus taken from the sea is sufficient to warm the air of the same region 1 up to a height of much more than 3,000 metres, since at high alt.i.tudes the air is subjected to less pressure, and consequently a cubic metre there contains less air than at the sea-level. But it is not a depth of 1 metre of the Gulf Stream that has been cooled 1 between these two sections; it is a depth of about 500 metres or more, and it has been cooled between 1 and 2 C. It will thus be easily understood that this loss of warmth from the Gulf Stream must have a profound influence on the temperature of the air over a wide area; we see how it comes about that warm currents like this are capable of rendering the climate of countries so much milder, as is the case in Europe; and we see further how comparatively slight variations in the temperature of the current from year to year must bring about considerable variations in the climate; and how we must be in a position to predict these latter changes when the temperature of the currents becomes the object of extensive and continuous investigation. It may be hoped that this is enough to show that far-reaching problems are here in question.

The salinity of the Gulf Stream water decreases considerably between the Fram's southern and northern sections. While in the former it was in great part between 35.4 and 35.5 per mille, in the latter it is throughout not much more than 35.3 per mille. In this section, also, the waters of the Gulf Stream are divided by an acc.u.mulation of less salt and somewhat colder bank water, which here lies over the Rockall Bank (Station 16). On the west side of this bank there is again (Station 15) salter and warmer Gulf Stream water, though not quite so warm as on the east. From the Frithjof section, a little farther south, it appears that this western volume of Gulf Stream water is comparatively small. The investigations of the Fram and the Frithjof show that the part of the Gulf Stream which penetrates into the Norwegian Sea comes in the main through the Rockall Channel, between the Rockall Bank and the bank to the west of the British Isles; its width in this region is thus considerably less than was usually supposed. Evidently this is largely due to the influence of the earth's rotation, whereby currents in the northern hemisphere are deflected to the right, to a greater degree the farther north they run. In this way the ocean currents, especially in northern lat.i.tudes, are forced against banks and coasts lying to the right of them, and frequently follow the edges, where the coast banks slope down to the deep. The conclusion given above, that the Gulf Stream comes through the Rockall Channel, is of importance to future investigations; it shows that an annual investigation of the water of this channel would certainly contribute in a valuable way to the understanding of the variations of the climate of Western Europe.

We shall not dwell at greater length here on the results of the Fram's oceanographical investigations in 1910. Only when the observations then collected, as well as those of the Frithjof's and Michael Sars's voyages, have been fully worked out shall we be able to make a complete survey of what has been accomplished.

Investigations in the South Atlantic, June to August, 1911.

In the South Atlantic we have the southward Brazil Current on the American side, and the northward Benguela Current on the African side. In the southern part of the ocean there is a wide current flowing from west to east in the west wind belt. And in its northern part, immediately south of the Equator, the South Equatorial Current flows from east to west. We have thus in the South Atlantic a vast circle of currents, with a motion contrary to that of the hands of a clock. The Fram expedition has now made two full sections across the central part of the South Atlantic; these sections take in both the Brazil Current and the Benguela Current, and they lie between the eastward current on the south and the westward current on the north. This is the first time that such complete sections have been obtained between South America and Africa in this part of the ocean. And no doubt a larger number of stations were taken on the Fram's voyage than have been taken -- with the same amount of detail -- in the whole South Atlantic by all previous expeditions put together.

When the Fram left Buenos Aires in June, 1911, the expedition went eastward through the Brazil Current. The first station was taken in lat. 36 18' S. and long. 43 15' W.; this was on June 17. Her course was then north-east or east until Station 32 in lat. 20 30'

S. and long. 8 10' E.; this station lay in the Benguela Current, about 800 miles from the coast of Africa, and it was taken on July 22. From there she went in a gentle curve

[Fig. 5 and caption]

past St. Helena and Trinidad back to America. The last station (No. 60) was taken on August 19 in the Brazil Current in lat. 24 39' S. and about long. 40 W.; this station lay about 200 miles south-east of Rio de Janeiro.

There was an average distance of 100 nautical miles between one station and the next. At nearly all the stations investigations were made at the following depths: surface, 5, 10, 25, 50, 100, 150, 200, 250, 300, 400, 500, 750, and 1,000 metres (2.7, 5.4, 13.6, 27.2, 54.5, 81.7, 109, 136.2, 163.5, 218, 272.5, and 545 fathoms). At one or two of the stations observations were also taken at 1,500 and 2,000 metres (817.5 and 1,090 fathoms).

The investigations were thus carried out from about the middle of July to the middle of August, in that part of the southern winter which corresponds to the period between the middle of

[Fig. 6]

Fig. 6. -- Currents in the South Atlantic (June -- August, 1911).

December and the middle of February in the northern hemisphere We must first see what the conditions were on the surface in those regions in the middle of the winter of 1911.

It must be remembered that the currents on the two sides of the ocean flow in opposite directions. Along the coast of Africa, we have the Benguela Current, flowing from south to north; on the American side the Brazil Current flows from the tropics southward. The former current is therefore comparatively cold and the latter comparatively warm. This is clearly seen on the chart, which shows the distribution of temperatures and salinities on the surface. In lat. 20 S. it was only about 17 C. off the African coast, while it was about 23 C. off the coast of Brazil.

The salinity depends on the relation between evaporation and the addition of fresh water. The Benguela Current comes from

[Fig. 7]

Fig. 7. -- Salinities and Temperatures at the Surface in the South Atlantic (June -- August, 1911) regions where the salinity is comparatively low; this is due to the acquisition of fresh water in the Antarctic Ocean, where the evaporation from the surface is small and the precipitation comparatively large. A part of this fresh water is also acquired by the sea in the form of icebergs from the Antarctic Continent. These icebergs melt as they drift about the sea.

Immediately off the African coast there is a belt where the salinity is under 35 per mille on the surface; farther out in the Benguela Current the salinity is for the most part between 35 and 36 per mille. As the water is carried northward by the current, evaporation becomes greater and greater; the air becomes comparatively warm and dry. Thereby the salinity is raised. The Benguela Current is then continued westward in the South Equatorial Current; a part of this afterwards turns to the north-west, and crosses the Equator into the North Atlantic, where it joins the North Equatorial Current. This part must thus pa.s.s through the belt of calms in the tropics. In this region falls of rain occur, heavy enough to decrease the surface salinity again. But the other part of the South Equatorial Current turns southward along the coast of Brazil, and is then given the name of the Brazil Current. The volume of water that pa.s.ses this way receives at first only small additions of precipitation; the air is so dry and warm in this region that the salinity on the surface rises to over 37 per mille. This will be clearly seen on the chart; the saltest water in the whole South Atlantic is found in the northern part of the Brazil Current. Farther to the south in this current the salinity decreases again, as the water is there mixed with fresher water from the South. The River La Plata sends out enormous quant.i.ties of fresh water into the ocean. Most of this goes northward, on account of the earth's rotation; the effect of this is, of course, to deflect the currents of the southern hemisphere to the left, and those of the northern hemisphere to the right. Besides the water from the River La Plata, there is a current flowing northward along the coast of Patagonia -- namely, the Falkland Current. Like the Benguela Current, it brings water with lower salinities than those of the waters farther north; therefore, in proportion as the salt water of the Brazil Current is mixed with the water from the River La Plata and the Falkland Current, its salinity decreases. These various conditions give the explanation of the distribution of salinity and temperature that is seen in the chart.

Between the two long lines of section there is a distance of between ten and fifteen degrees of lat.i.tude. There is, therefore, a considerable difference in temperature. In the southern section the average surface temperature at Stations 1 to 26 (June 17 to July 17) was 17.9 C.; in the northern section at Stations 36 to 60 (July 26 to August 19) it was 21.6 C. There was thus a difference of 3.7 C. If all the stations had been taken simultaneously, the difference would have been somewhat greater; the northern section was, of course, taken later in the winter, and the temperatures were therefore proportionally lower than in the southern section. The difference corresponds fairly accurately with that which Kr:ummel has calculated from previous observations.

We must now look at the conditions below the surface in that part of the South Atlantic which was investigated by the Fram Expedition.

The observations show in the first place that both temperatures and salinities at every one of the stations give the same values from the surface downward to somewhere between 75 and 150 metres (40.8 and 81.7 fathoms). This equalization of temperature and salinity is due to the vertical currents produced by cooling in winter; we shall return to it later. But below these depths the temperatures and salinities decrease rather rapidly for some distance.

The conditions of temperature at 400 metres (218 fathoms) below the surface are shown in the next little chart. This chart is based on the Fram Expedition, and, as regards the other parts of the ocean, on Schott's comparison of the results of previous expeditions. It will be seen that the Fram's observations agree very well with previous soundings, but are much more detailed.

The chart shows clearly that it is much warmer at 400 metres (218 fathoms) in the central part of the South Atlantic than either farther north -- nearer the Equator -- or farther south. On the Equator there is a fairly large area where the temperature is only 7 or 8 C. at 400 metres, whereas in lats. 2O to 30 S. there are large regions where it is above 12 C.; sometimes above 13 C., or even 14C. South of lat. 30 S. the temperature decreases again rapidly; in the chart no lines are drawn for temperatures below 8 C., as we have not sufficient observations to show the course of these lines properly. But we know that the temperature at 400 metres sinks to about 0 C. in the Antarctic Ocean.

[Fig. 8]

Fig. 8. -- Temperatures (Centigrade) at a Depth of 400 Metres (218 Fathoms).

At these depths, then, we find the warmest water within the region investigated by the Fram. If we now compare the distribution of temperature at 400 metres with the chart of currents in the South Atlantic, we see that the warm region lies in the centre of the great circulation of which mention was made above. We see that there are high temperatures on the left-hand side of the currents, and low on the right-hand side. This, again, is an effect of the earth's rotation, for the high temperatures mean as a rule that the water is comparatively light, and the low that it is comparatively heavy. Now, the effect of the earth's rotation in the southern hemisphere is that the light (warm) water from above is forced somewhat down on the left-hand side of the current, and that the heavy (cold) water from below is raised somewhat. In the northern hemisphere the contrary is the case. This explains the cold water at a depth of 400 metres on the Equator; it also explains the fact that the water immediately off the coasts of Africa and South America is considerably colder than farther out in the ocean. We now have data for studying the relation between the currents and the distribution of warmth in the volumes of water in a way which affords valuable information as to the movements themselves. The material collected by the Fram will doubtless be of considerable importance in this way when it has been finally worked out.

Below 400 metres (218 fathoms) the temperature further decreases everywhere in the South Atlantic, at first rapidly to a depth between 500 and 1,000 metres (272.5 and 545 fathoms), afterwards very slowly. It is possible, however, that at the greatest depths it rises a little again, but this will only be a question of hundredths, or, in any case, very few tenths of a degree.

It is known from previous investigations in the South Atlantic, that the waters at the greatest depths, several thousand metres below the surface, have a temperature of between 0 and 3 C. Along the whole Atlantic, from the extreme north (near Iceland) to the extreme south, there runs a ridge about half-way between Europe and Africa on the one side, and the two American continents on the other. A little to the north of the Equator there is a slight elevation across the ocean floor between South America and Africa. Farther south (between lats. 25 and 35 S.) another irregular ridge runs across between these continents. We therefore have four deep regions in the South Atlantic, two on the west (the Brazilian Deep and the Argentine Deep) and two on the east (the West African Deep and the South African Deep). Now it has been found that the "bottom water" in these great deeps -- the bottom lies more than 5,000 metres (2,725 fathoms) below the surface -- is not always the same. In the two western deeps, off South America, the temperature is only a little above 0 C. We find about the same temperatures in the South African Deep, and farther eastward in a belt that is continued round the whole earth. To the south, between this belt and Antarctica, the temperature of the great deeps is much lower, below 0 C. But in the West African Deep the temperature is about 2 C. higher; we find there the same temperatures of between 2 and 2.5 C. as are found everywhere in the deepest parts of the North Atlantic. The explanation of this must be that the bottom water in the western part of the South Atlantic comes from the south, while in the north-eastern part it comes from the north. This is connected with the earth's rotation, which has a tendency to deflect currents to the left in the southern hemisphere. The bottom water coming from the south goes to the left -- that is, to the South American side; that which comes from the north also goes to the left -- that is, to the African side.

The salinity also decreases from the surface downward to 600 to 800 metres (about 300 to 400 fathoms), where it is only a little over 34 per mille, but under 34.5 per mille; lower down it rises to about 34.7 per mille in the bottom water that comes from the south, and to about 34.9 per mille in that which comes from the North Atlantic.

We mentioned that the Benguela Current is colder and less salt at the surface than the Brazil Current. The same thing is found in those parts of the currents that lie below the surface. This is clearly shown in Fig. 9, which gives the distribution of temperature at Station 32 in the Benguela Current, and at Station 60 in the Brazil Current; at the various depths down to 500 metres (272.5 fathoms) it was between 5 and 7 C. colder in the former than in the latter. Deeper down the difference becomes less, and at 1,000 metres (545 fathoms) there was only a difference of one or two tenths of a degree.

Fig. 10 shows a corresponding difference in salinities; in the first 200 metres below the surface the water was about

[Fig. 9.]